A  AuSSt;  fjis?"'  )      Relative  Activity  of  Digitalis.  585 
Table  III  further  demonstrates  that  the  hairs  of  digitalis  are 
practically  devoid  of  glucosidal  principles.  This  is  further  substan- 
tiated by  the  fact  that  a  microscopical  examination  of  the  fine  sitt- 
ings revealed  the  characteristic  hairs  of  digitalis  as  by  far  the  larger 
portion  of  all  tissue  present.  Again  no  qualitative  gludosidal  test 
could  be  obtained  from  an  infusion  of  this  sample. 
Tables  IV,  V,  VI  give  in  detail  the  results  of  the  further  ex- 
amination of  the  three  separated  portions,  i.  e.,  (a)  select  cleaned 
digitalis,  (&)  petioles,  and  (c)  number  50  dirt  siftings.  These  por- 
tions were  examined  by  a  modification  of  the  Hatcher  Ouabain 
Method  suggested  by  Hatcher  and  used  by  Rowntree  and  Macht.  A 
M.  L.  D.  of  100  milligrammes  of  drug  per  kilogramme  of  cat, 
injected  intravenously,  is  generally  accepted  as  the  average  lethal 
dose  for  digitalis.  Table  IV  shows  that  the  cat  unit  when  ex- 
amined by  the  method  above  mentioned  is  equal  to  87.62  milli- 
grammes of  select  cleaned  digitalis  leaves,  which  is  12.38  per  cent, 
more  toxic  than  the  average  drug.  Obviously  it  is  impractical  to 
compare  the  relative  value  of  the  clean  digitalis  as  determined  by  the 
guinea  pig  and  cat  methods  of  assay  unless  the  respective  standards 
have  a  known  relative  value.  It  is  our  opinion  that  .2  milligrammes 
of  digitalis  per  gramme  of  guinea  pig  is  the  approximate  equivalent 
of  100  milligrammes  per  Kg.  of  cat.  Our  results  show  that  .175 
milligramme  of  drug  per  gramme  of  guinea  pig  represents  the  ap- 
proximate equivalent  of  87.62  milligrammes  of  drug  per  Kg.  of  cat. 
On  this  basis  both  methods  of  assay  indicate  that  the  specially  cleaned 
Minnesota  digitalis  was  about  12  per  cent,  more  active  than  the 
average  sample  of  digitalis. 
A  comparison  of  the  results  presented  in  Table  IV  with  those 
shown  in  Table  V  indicates  that  the  petioles  were  about  one  fourth 
as  active  as  the  cleaned  drug.  This  is  in  close  accordance  with  the 
results  obtained  by  the  guinea-pig  method. 
Table  VI  relates  to  the  examination  of  the  fine  dirt  siftings  the 
same  as  reported  upon  by  the  guinea-pig  method  in  Table  III.  While 
the  results  seem  to  be  somewhat  discordant,  in  general  they  corrob- 
orate with  Table  III.  Eliminating  cats  number  one,  six,  three  and 
seven,  the  variation  is  not  more  than  would  be  expected  considering 
the  unusual  symptoms  exhibited  by  the  guinea  pigs,  and  the  excep- 
tional symptoms  which  were  also  evidenced  in  the  cats.  All  the  cats 
exhibited  marked  peristalsis,  defagation  taking  place  in  three  of  the 
animals.   With  one  exception  the  cats  showed  the  usual  decrease  in 
